Today's Elites

Sunday, September 11, 2011

Game Theory and Oligarchy

It has always been my reading that among the means of suppression of natural human creativity by an oligarchic clique is to in effect bottle it up in a sort of labyrinth of "games." The writing of Pascal on the subject of why the ruler must be amused so that the horror of his own Ozymandian downfall need not be contemplated. It is ironically significant then that Pascal's work on probabilities of game playing led to Leibniz' development of a singularly powerful tool for improving human productivity: the calculus. Today we have Bill Gates, "artificial intelligence," and video games. (Oh, and Obama's quaint fixation on basketball.)

Now, it is not as if there is some "vast conspiracy" out there to capture the mind of man and rob it of a useful application. Rather, it is a worldview that produces a certain type of output. It is through the slavish acceptance of certain "common sense" axioms that we as a society are ensnared.

Such as: "money rules the world." Why? One might ask. But to ask such is to deny the certainties of one's senses. Why, it is the way of all flesh, thou dolt! It would be like denying that you are ruled by the principles of pleasure and pain. Like denying the right of Caesar, or King George, or the George Bushes...
And now dear reader enough folderol. Behold the latest very learned essay on games and artificial intelligence and tremble:

Measuring Intelligence through Games

Artificial general intelligence (AGI) refers to research aimed at tackling the full problem of artificial intelligence, that is, create truly intelligent agents. This sets it apart from most AI research which aims at solving relatively narrow domains, such as character recognition, motion planning, or increasing player satisfaction in games. But how do we know when an agent is truly intelligent? A common point of reference in the AGI community is Legg and Hutter's formal definition of universal intelligence, which has the appeal of simplicity and generality but is unfortunately incomputable. Games of various kinds are commonly used as benchmarks for "narrow" AI research, as they are considered to have many important properties. We argue that many of these properties carry over to the testing of general intelligence as well. We then sketch how such testing could practically be carried out. The central part of this sketch is an extension of universal intelligence to deal with finite time, and the use of sampling of the space of games expressed in a suitably biased game description language.


My, my...Very learned, indeed...

No comments:

Post a Comment

Blog Archive